Search NASA⌕ Search

SEARCH · Search NASA

Results for “treatment learning”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Adapting Reinforcement Learning Treatment Policies Using Limited Data to Personalize Critical Care

We report reinforcement learning (RL) demonstrates promise for developing effective treatment policies in critical care settings. However, existing RL methods often require large and comprehensive patient data sets and do not readily lend themselves to settings in which certain patient subpopulations are severely underrepresented. In this study, we develop a new method, noisy Bayesian policy updates (NBPU), for selecting high-performing reinforcement learning–based treatment policies for underrepresented patient subpopulations using limited observations. Our method uses variational inference to learn a probability distribution over treatment policies based on a reference patient subpopulation for which sufficient data are available. It then exploits limited data from an underrepresented patient subpopulation to update this probability distribution and adapts its recommendations to this subpopulation. We demonstrate our method’s utility on a data set of ICU patients receiving intravenous blood anticoagulant medication. Our results show that NBPU outperforms state-of-the-art methods in terms of both selecting effective treatment policies for patients with nontypical clinical characteristics and predicting the corresponding policies’ performance for these patients.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

A physics-constrained deep learning treatment of runaway electron dynamics

An adjoint formulation leveraging a physics-informed neural network (PINN) is employed to advance the density moment of a runaway electron (RE) distribution forward in time. A distinguishing feature of this approach is that once the adjoint problem is solved, its solution can be used to project the RE density forward in time for an arbitrary initial momentum space distribution of REs. Furthermore, by employing a PINN, a parametric solution to the adjoint problem can be learned. Thus, once trained, this adjoint-deep learning framework is able to efficiently project the RE density forward in time across various plasma conditions while still including a fully kinetic description of RE dynamics. As an example application, the temporal evolution of the density of primary electrons is studied, with particular emphasis on evaluating the decay of a RE population when below threshold. Predictions from the adjoint-deep learning framework are found to be in good agreement with a traditional relativistic electron Fokker–Planck solver, for several distinct initial conditions, and across an array of physics parameters. Once trained, the PINN thus provides a means of generating RE density time histories with exceptionally low online execution time.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Open-source generation of sigma profiles: impact of quantum chemistry and solvation treatment on machine learning performance

The combination of machine learning (ML) models with chemistry-related tasks requires the description of molecular structures in a machine-readable way. The nature of these so-called molecular descriptors has a direct and major impact on the performance of ML models and remains an open problem in the field. Structural descriptors like SMILES strings or molecular graphs lack size-independence and can be memory intensive. Machine-learned descriptors can be of low dimensionality and constant size but lack physical significance and human interpretability. Sigma profiles, which are unnormalized histograms of the surface charge distributions of solvated molecules, combine physical significance with low dimensionality and size-independence, making them a suitable candidate for a universal molecular descriptor. However, their widespread adoption in ML applications requires open access to sigma profile generation, which is currently not available. This work details the development of OpenSPGen – an open-source tool for generating sigma profiles. Also presented are studies on the effect of different settings on the efficacy of the generated sigma profiles at predicting thermophysical material properties when used as inputs to a Gaussian process as a simple surrogate ML model. We find that a higher level of theory does not translate to more accurate results. We also provide further recommendations for sigma profile calculation and use in ML models.

Salih, Fathya Y. M. [University of Notre Dame, IN ↗
DOE OSTI2020

Enhancement of microbiome management by machine learning for biological wastewater treatment

Here we propose to develop microbiome-based machine learning models to predict the response of biological wastewater treatment systems to environmental or operational disturbances or to design specific microbiomes to achieve a desired system function. These machine learning models can be used to enhance the stability of microbiome-based biological systems and warn against the failure of these systems.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Evaluation of Hydrogen Gettering Rates Correlated to Surface Composition and Texture of Nickel-Plated Zircaloy Getters of Different Heat Treatment Procedures

Coatings of metal specimens are known to have an impact on hydrogen gettering (hydrogen absorption). The coating can have one or more functions, such as enhancing gettering, preventing gettering and/or preventing oxidation of the metal substrate. It is known that contaminants and surface texture can impact hydrogen gettering/absorption performance, but has not previously been thoroughly explored. This study evaluated the role of different post-plating heat treatments of nickel-plated zircaloy-4 getters (NPGs) and the role of the heat treatments on gettering rates, surface composition and texture. Nickel plating is applied to prevent oxidation of the Zircaloy-4 surface and also enhances gettering. The nickel plating must be heat treated before desirable gettering can occur. Our NPG getters with historically known satisfying performance were pre-heat treated in air followed by activation heat treatment in a vacuum at a higher temperature. In this study, we were interested in finding out if both heat treatment steps were necessary to obtain a desirable gettering performance, or if one step could be omitted. XPS analysis showed that if the nickel surface is not heat treated before bonding the nickel to the zirconium in the activation step, there will be carbon contaminants on the surface, which significantly reduces gettering. We studied the texture of Zircaloy-4 using SEM/EBSD to compare NPGs with both heat treatment steps with NPGs that had no post-plating heat treatment to learn if the degree of cold work could be impacted by the heat treatment steps. We did not observe any differences in texture between them. We measured gettering rates of both pretreated and activated NPGs and NPGs that had been activated without first being pre-heat treated. We found that the NPGs without the first post-plating heating step had up to a seven times slower gettering rate and obtained higher plateau pressures due to the contaminated surface. Thus, the pre-heat treatment in air before activation is necessary to avoid slower gettering rates and higher plateau pressures.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Perspectives on Machine Learning-assisted Plasma Medicine: Towards Automated Plasma Treatment

Cold atmospheric plasmas (CAPs) have shown great promise for medical applications through their synergistic chemical, electrical, and thermal effects, which can induce therapeutic outcomes. However, safe and reproducible plasma treatment of complex biological surfaces poses a major hurdle to the widespread adoption of CAPs for medical applications. Predictive modeling of the mutual interactions between the plasma and biological surfaces and, thus, systematic approaches to quantify and predict plasma treatment outcomes remain largely elusive due to the lack of mechanistic understanding of plasma-surface interactions that can span across vastly different length-and time-scales. In addition, real-time sensing capabilities in biomedical CAP devices are often limited, which can be detrimental to plasma treatment due to the intrinsic plasma and surface variability during the treatment, as well as sensitivity to external perturbations. All of these challenges can make reproducible and effective plasma treatment of biological surfaces difficult to realize, which is further compounded by errors due to human operation of hand-held CAP devices. Machine learning and data-driven approaches can be particularly useful in addressing these challenges in three major ways: (i) data-driven modeling of hard-to-model plasma-surface interactions and plasma treatment outcomes; (ii) learning data analytics for plasma and surface diagnostics in real-time; and (iii) developing predictive controllers that enable reliable and effective CAP treatments. Furthermore, this paper discusses the promise of machine learning to accelerate plasma medicine research in these areas, toward machine learning-assisted and automated CAP treatment of complex biological surfaces.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2021

A cross-study analysis of drug response prediction in cancer cell lines

Abstract To enable personalized cancer treatment, machine learning models have been developed to predict drug response as a function of tumor and drug features. However, most algorithm development efforts have relied on cross-validation within a single study to assess model accuracy. While an essential first step, cross-validation within a biological data set typically provides an overly optimistic estimate of the prediction performance on independent test sets. To provide a more rigorous assessment of model generalizability between different studies, we use machine learning to analyze five publicly available cell line-based data sets: National Cancer Institute 60, ancer Therapeutics Response Portal (CTRP), Genomics of Drug Sensitivity in Cancer, Cancer Cell Line Encyclopedia and Genentech Cell Line Screening Initiative (gCSI). Based on observed experimental variability across studies, we explore estimates of prediction upper bounds. We report performance results of a variety of machine learning models, with a multitasking deep neural network achieving the best cross-study generalizability. By multiple measures, models trained on CTRP yield the most accurate predictions on the remaining testing data, and gCSI is the most predictable among the cell line data sets included in this study. With these experiments and further simulations on partial data, two lessons emerge: (1) differences in viability assays can limit model generalizability across studies and (2) drug diversity, more than tumor diversity, is crucial for raising model generalizability in preclinical screening.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Machine learning models of intermittent operation of RO wellhead water treatment for salinity reduction and nitrate removal

Machine learning models were developed for intermittent multi-mode operation of a wellhead reverse osmosis water purification and desalination system to predict salt passage, nitrate passage, and permeate flux. The models, based on long short-term memory (LSTM) recurrent neural network (RNN) architecture, included an attention mechanism to increase model performance in proximity of the regulatory limit for nitrate. Training and testing of the models for the Startup, Production, Shutdown and Flushing operational modes were based on operational data (consisting of 22 process variables per data sample) acquired every 2–5 s over a six-month period. The significant sets of model input attributes for the different operational modes were assessed via Spearman ranking correlation, Self-Organizing Map (SOM) analysis and feed forward feature selection (FFFS). Although the variability of nitrate passage, salt passage and permeate flux was significant over the four operational modes, prediction performance for the three outcomes were with R2 and Average Absolute Relative Error (AARE) of 0.78–0.95 and 2.96–6.16 %, respectively. Model updates post membrane elements replacement demonstrated similar levels of prediction accuracy. The study results suggest that there is merit in exploring the utility of multi-mode models for sensor fault detection, data imputation, and for potential use in model-predictive control.

Intermittent RO operation↗
DOE OSTI2022

Prototyping a self-learning digital twin platform for personalized treatment in melanoma patients (Final Report)

We completed all the Milestones and disseminated results at the 2021 Computational Approaches for Cancer Workshop (CAFCW21) at the SuperComputing21 conference. As part of Milestone 1, we developed and implemented a prototype digital twin of metastatic melanoma patients, performed quality checks on the code and released it as open source. The work has directly driven major grant proposals by the team on computational digital twins and supporting software infrastructure, including a successful fellowship application by the PI. Work was lead by Paul Macklin (Contact PI, Indiana University), in collaboration with Ilya Shmulevich (PI, Institute for Systems Biology), Tina Hernandez-Boussard (PI, Stanford Univ.), Jeffrey Bryan (Co-I, University of Missouri), Snigdhansu Chatterjee (Co-I, University of Minnesota), and Mohammad Fallahi-Sichani (Co-I, U. of Virginia). Postdoctoral student Heber L. Rocha (Macklin lab, IU) performed intensive computational work, in collaboration with Senior Research Scientist Boris Aguilar (Shmulevich lab, ISB).

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Optimizing Individualized Treatment Planning for Parkinson’s Disease Using Deep Reinforcement Learning

More than one million people currently live with Parkinson's Disease (PD) in the U.S. alone. Medications, such as levodopa, can help manage PD symptoms. However, medication treatment planning is generally based on patient history and limited interaction between physicians and patients during office visits. This limits the extent of benefit that may be derived from the treatment as disease/patient characteristics are generally non-stationary. Wearable sensors that provide continuous monitoring of various symptoms, such as bradykinesia and dyskinesia, can enhance symptom management. However, using such data to overhaul the current static medication treatment planning approach and prescribe personalized medication timing and dosage that accounts for patient/care-giver/physician feedback/preferences remains an open question. We develop a model to prescribe timing and dosage of medications, given the motor fluctuation data collected using wearable sensors in real-time. We solve the resulting model using deep reinforcement learning (DRL). The prescribed policy determines the optimal treatment plan that minimizes patient's symptoms. Our results show that the model-prescribed policy outperforms the static a priori treatment plan in improving patients' symptoms, providing a proof-of-concept that DRL can augment medical decision making for treatment planning of chronic disease patients.

Watts, Jeremy↗
DOE OSTI2023

Deep learning methods for drug response prediction in cancer: Predominant and emerging trends

Cancer claims millions of lives yearly worldwide. While many therapies have been made available in recent years, by in large cancer remains unsolved. Exploiting computational predictive models to study and treat cancer holds great promise in improving drug development and personalized design of treatment plans, ultimately suppressing tumors, alleviating suffering, and prolonging lives of patients. A wave of recent papers demonstrates promising results in predicting cancer response to drug treatments while utilizing deep learning methods. These papers investigate diverse data representations, neural network architectures, learning methodologies, and evaluations schemes. However, deciphering promising predominant and emerging trends is difficult due to the variety of explored methods and lack of standardized framework for comparing drug response prediction models. To obtain a comprehensive landscape of deep learning methods, we conducted an extensive search and analysis of deep learning models that predict the response to single drug treatments. A total of 61 deep learning-based models have been curated, and summary plots were generated. Based on the analysis, observable patterns and prevalence of methods have been revealed. This review allows to better understand the current state of the field and identify major challenges and promising solution paths.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2021

A vision of immuno-oncology: the Siena think tank of the Italian Network for Tumor Biotherapy (NIBIT) foundation

The yearly Think Tank Meeting of the Italian Network for Tumor Biotherapy (NIBIT) Foundation, brings together in Siena, Tuscany (Italy), experts in immuno-oncology to review the learnings from current immunotherapy treatments, and to propose new pre-clinical and clinical investigations in selected research areas. While immunotherapies in non-small cell lung cancer and melanoma led to practice changing therapies, the same therapies had only modest benefit for patients with other malignancies, such as mesothelioma and glioblastoma. One way to improve on current immunotherapies is to alter the sequence of each combination agent. Matching the immunotherapy to the host’s immune response may thus improve the activity of the current treatments. A second approach is to combine current immunotherapies with novel agents targeting complementary mechanisms. Identifying the appropriate novel agents may require different approaches than the traditional laboratory-based discovery work. For example, artificial intelligence-based research may help focusing the search for innovative and most promising combination partners. Novel immunotherapies are needed in cancer patients with resistance to or relapse after current immunotherapeutic drugs. Such new treatments may include targeted agents or monoclonal antibodies to overcome the immune-suppressive tumor microenvironment. The mode of combining the novel treatments, including vaccines, needs to be matched to the patient’s immune status for achieving the maximum benefit. In this scenario, specific attention should be also paid nowadays to the immune intersection between COVID-19 and cancer.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

Aging heat treatment design for Haynes 282 made by wire-feed additive manufacturing using high-throughput experiments and interpretable machine learning

Wire-feed additive manufacturing (WFAM) produces superalloys with complex thermal cycles and unique microstructures, often requiring optimized heat treatments. To address this challenge, we present a hybrid approach that combines high-throughput experiments, precipitation simulation, and machine learning to design effective aging conditions for the WFAM Haynes 282 superalloy. Our results demonstrate that the γ’ radius is the critical microstructural feature for strengthening Haynes 282 during post-heat treatment compared with the matrix composition and γ’ volume fraction. New aging conditions at 770°C for 50 hours and 730°C for 200 hours were discovered based on the machine learning model and were applied to enhance yield strength, bringing it on par with the wrought counterpart. This approach has significant implications for future AM alloy production, enabling more efficient and effective heat treatment design to achieve desired properties.

CALPHAD↗
DOE OSTI2026

Molecular pathways for learning in the single-cell Stentor coeruleus

The single-cell Stentor coeruleus contracts in response to mechanical taps but habituates and learns to ignore the taps after repeated stimulation. Here, we explored the molecular changes that occur during the formation of this cellular memory in order to improve our understanding of non-synaptic learning. We impaired cellular protein synthesis with cycloheximide and puromycin and found that, contrary to the effects of such treatments on metazoa, these drugs accelerate habituation and prolong memory retention in Stentor . Exploratory proteomic and transcriptomic analyses identified candidate proteins and genes that changed over the course of habituation and response recovery, pointing toward the regulation of Stentor learning by calcium signaling and protein phosphorylation. Building on these results, we found that using RNA interference to knock down the calcium-binding, EF-hand domain-containing protein SteCoe_6763 accelerated habituation. Furthermore, increased extracellular calcium improved Stentor learning, while treatment with kinase and phosphatase inhibitors impaired learning. In particular, KN-93, a drug known to inhibit calcium/calmodulin-dependent kinase II and voltage-gated calcium channels, decreased both the rate and extent of habituation in Stentor , similar to its effects on learning in metazoa. We also discovered that habituation memory can be maintained in progeny following cell division. Taken together, these results suggest that response recovery in Stentor requires new protein synthesis and that memory formation involves the modification of delocalized mechanoreceptors by phosphorylation and calcium signaling. This is consistent with our previous model of Stentor learning, in which habituation occurs through the inactivation of cell-surface receptors.

CaMKII↗